The aims of this study are to 1) examine the relationships between various indices of rice quality and agro-climatological elements, and to 2) contribute to the evaluation of projecting climate change impact on rice quality in Japan. We used a large collection of rice quality dataset from the Crop Survey database provided ad hoc by the Ministry of Agriculture, Forestry and Fisheries of Japan. Firstly, we conducted statistical analysis to determine the interrelationships among the various indices of rice quality such as the percentages of the first-grade rice, undamaged rice and chalky grain, then also analyzed their relationships with surface meteorological elements and their composite indices. The results showed that the indices of rice qualities were well correlated with a heat stress index HD_m26, that is a cumulative deviation of daily averaged surface air temperature from a base of 26 °C during the 20days after heading. Especially, the percentage of chalky grain was better explained in relation with the HD_m26 than that of simple 20 days-averaged air temperature during the corresponding period. In addition, the Quantile Regression analysis was applied to regress the percentage of chalky grain based on the HD_m26. Although the result showed the significant linear, quadratic and cubic curves regression in each percentile values of the chalky grain with the zone of HD_m26 values, varietal differences were perceived in these statistical relationships.
